CN117755354A - High-speed railway line disease early warning method and device and computer equipment - Google Patents
High-speed railway line disease early warning method and device and computer equipment Download PDFInfo
- Publication number
- CN117755354A CN117755354A CN202311565787.1A CN202311565787A CN117755354A CN 117755354 A CN117755354 A CN 117755354A CN 202311565787 A CN202311565787 A CN 202311565787A CN 117755354 A CN117755354 A CN 117755354A
- Authority
- CN
- China
- Prior art keywords
- overrun
- data
- disease
- point cluster
- acceleration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 201000010099 disease Diseases 0.000 title claims abstract description 181
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 181
- 238000000034 method Methods 0.000 title claims abstract description 50
- 230000003321 amplification Effects 0.000 claims abstract description 30
- 238000003199 nucleic acid amplification method Methods 0.000 claims abstract description 30
- 238000004458 analytical method Methods 0.000 claims abstract description 23
- 238000007621 cluster analysis Methods 0.000 claims abstract description 19
- 238000011161 development Methods 0.000 claims abstract description 16
- 230000001133 acceleration Effects 0.000 claims description 109
- 238000012423 maintenance Methods 0.000 claims description 21
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 6
- 230000003247 decreasing effect Effects 0.000 claims description 4
- 238000007689 inspection Methods 0.000 abstract description 20
- 230000000694 effects Effects 0.000 abstract description 2
- 238000004422 calculation algorithm Methods 0.000 description 13
- 238000011835 investigation Methods 0.000 description 10
- 229910000831 Steel Inorganic materials 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 4
- 239000010959 steel Substances 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000005299 abrasion Methods 0.000 description 2
- 230000001174 ascending effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 208000037920 primary disease Diseases 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005452 bending Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Abstract
The application relates to a high-speed railway line disease early warning method, a device and computer equipment, wherein the method comprises the steps of obtaining overrun data of a target train in a target time period; inputting the overrun data into a pre-established driving speed-acceleration amplification model to obtain an overrun result data set; performing density cluster analysis on the overrun result data set to obtain a first overrun data point cluster; performing development trend analysis on the first overrun data point cluster to obtain a second overrun data point cluster and disease grades corresponding to the second overrun data point cluster; and carrying out early warning according to the second overrun data point cluster and the disease grade corresponding to the second overrun data point cluster. The method has the effects of more timely line disease discovery, more accurate line disease position positioning, reduction of manual inspection cost, and reduction of false inspection and missing inspection phenomena.
Description
Technical Field
The application relates to the technical field of rail transit, in particular to a high-speed railway line disease early warning method, a device and computer equipment.
Background
The track diseases refer to various damage, faults or abnormal conditions in track traffic systems such as railways, subways and the like, and can influence the running safety and the running efficiency of trains, and the existence of the track diseases can lead to unstable running of the trains, increased noise and shortened service life of the tracks. High-speed railways may cause very serious accidents once serious damage exists to the track due to the high running speed of the train.
In the related art, manual inspection mainly depends on skylight time (generally, the next half night), is limited by environment and time length, has long inspection interval time and short inspection operation time, and is difficult to discover diseases of a high-speed railway track in time.
Disclosure of Invention
In order to at least partially solve the technical problems, the application provides a high-speed railway line disease early warning method, a high-speed railway line disease early warning device and computer equipment.
In a first aspect, the present application provides a method for early warning a high-speed railway line disease, which adopts the following technical scheme:
the high-speed railway line disease early warning method is characterized by comprising the following steps of:
acquiring overrun data of a target train in a target time period, wherein the overrun data comprises acceleration information of the target train, and a work area, a line, a running category, detection time, mileage and running speed corresponding to the acceleration information;
Inputting the overrun data into a pre-established driving speed-acceleration amplification model to obtain an overrun result data set, wherein the overrun result data set is configured as a set of maximum acceleration reached by a target train when each line runs at the maximum driving speed;
performing density cluster analysis on the overrun result data set to obtain a first overrun data point cluster;
performing development trend analysis on the first overrun data point cluster to obtain a second overrun data point cluster and disease grades corresponding to the second overrun data point cluster;
and carrying out early warning according to the second overrun data point cluster and the disease grade corresponding to the second overrun data point cluster.
By adopting the technical proposal, the utility model has the advantages that,
the method comprises the steps that according to acceleration information, analysis of track local irregularity can be carried out, wherein the acceleration information is vibration acceleration of a target train, the vibration acceleration comprises horizontal vibration acceleration and vertical vibration acceleration, factors affecting the horizontal vibration acceleration of the target train comprise curves, continuous small directions of a turnout area, hard bends of a steel rail, joint support nozzles, track gauges, poor track gauge change rates, alternate uneven abrasion of straight sections of the steel rail and other diseases, factors affecting the vertical vibration acceleration of the target train comprise diseases such as poor track geometry, poor joint comprehensive state, serious track bed elasticity or uneven sections and the like, the horizontal vibration acceleration and the vertical vibration acceleration are generally divided into four grades according to the numerical value, early warning is needed when approaching three grades, three grades and four grades, and the obtained acceleration information can be subjected to analysis in subsequent steps to early warning on the disease condition of the track;
The specific position of the target train at a certain moment, namely the position of a certain section of track is analyzed in a local irregularity manner, wherein the work area can comprise a responsible unit and a unit number corresponding to the section of track, the line can comprise a line name and a line number corresponding to the section of track, the line respectively represents the running direction of the train on the section of track and comprises an ascending direction and a descending direction, the mileage represents the position of a mileage pile corresponding to the section of track, the running speed represents the running speed of the train on the section of track when data acquisition is performed, and the position of the target train and the disease degree of the section of track when the data acquisition is performed can be obtained more accurately by combining acceleration information;
when the speed of the target train is not fixed, the target train is possibly at a lower running speed, and the corresponding vibration acceleration is lower at the lower running speed, so that the disease condition of the track is difficult to reflect through the vibration acceleration, therefore, the acquired running speed and vibration acceleration are converted to the maximum running speed and the corresponding maximum vibration acceleration through a running speed-acceleration amplification model, and meanwhile, the data is more convenient to analyze after conversion;
Generally, a disease point affects a section of track around the disease point, that is, a plurality of overrun data points may be caused by the same disease point, and by dividing a plurality of overrun data points in a preset neighborhood range into a cluster of points, the approximate range of the disease point can be obtained, and an operation and maintenance person can conduct investigation in the corresponding range;
the development trend is mainly the development trend of vibration acceleration, the trend is different, may correspond to different diseases, diseases have been processed, disease serious conditions are different, etc., can also obtain the correspondent disease grade according to the concrete numerical value change condition of vibration acceleration, after excluding the conditions that the diseases have been processed, disease severity is lower, etc., the data point cluster of overrun obtained is the second overrun data point cluster;
the second overrun data point cluster can obtain a more accurate disease point range and severity, and can guide operation and maintenance personnel to go to the investigation treatment in time.
Optionally, the step of establishing the driving speed-acceleration amplification model includes:
acquiring first historical overrun data;
calculating a data base line according to the first historical overrun data to obtain a disease rule;
preprocessing the first historical overrun data to obtain a second historical data set;
Obtaining overrun data distribution and overrun data trend according to the disease rules and the second historical data set;
and generating the driving speed-acceleration amplification model according to the overrun data distribution and the overrun data trend.
By adopting the technical scheme, an accurate running speed-acceleration amplification model can be established according to the historical overrun data, so that the acquired running speed and vibration acceleration are converted during early warning, and the maximum vibration acceleration corresponding to the running of the target train at the maximum running speed is obtained;
the first historical overrun data can be selected according to actual conditions, such as historical overrun data of the last year or two years, a data base line can be calculated by manually analyzing historical vehicle shaking instrument data and operation maintenance data distribution conditions, trends and rules and combining the actual conditions of rail diseases, and finally overrun data with a certain proportion in a certain specific radius with the actual disease point as a circle center is obtained and corresponds to one disease point, namely a disease rule;
the preprocessing is mainly to acquire data of vibration acceleration in a certain section, such as data of vibration acceleration in the vicinity of three stages, in the first historical overrun data, under the condition that the vibration acceleration is low when the vehicle runs at the highest speed, the road section can be judged to have low disease degree, the processing is not needed temporarily, the higher vibration acceleration usually corresponds to the serious disease degree, the probability of the serious degree to continue to develop is also high, and therefore, after the preprocessing, a model can be built by more suitable data.
Optionally, the step of calculating a data baseline according to the first historical overrun data to obtain a disease rule includes:
obtaining a disease radius corresponding to a disease point according to the distribution condition, the trend and the rule of the historical vehicle shaking instrument data, the distribution condition, the trend and the rule of the operation maintenance data and the disease point data in the first historical overrun data;
and analyzing the correlation between the running speed and the acceleration in the first historical overrun data to obtain a running speed difference section, wherein the running speed difference section is configured such that the acceleration corresponding to a plurality of running speeds with running speed differences within the running speed difference section is the same.
By adopting the technical scheme, the method can obtain more accurate disease radius corresponding to the disease point according to the historical data, so that density cluster analysis can be carried out in the subsequent step, and the calculation amount in the process of establishing a model and subsequent early warning can be reduced through the driving speed difference section, so that the early warning speed is improved.
Optionally, the step of preprocessing the first historical overrun data to obtain a second historical data set includes calculating an overrun interval according to a preset acceleration level and an acceleration overrun range, and selecting data of overrun acceleration of each line in the overrun interval according to the overrun interval and the first historical overrun data to obtain the second historical data set; the step of obtaining the overrun data distribution and the overrun data trend according to the disease rule and the second historical data set comprises the steps of calculating model parameters corresponding to each line according to the disease rule and the second historical data set, and taking the model parameters as the overrun data distribution and the overrun data trend, wherein the model parameters comprise a slope, an intercept, a standard deviation of a prediction error, a t statistic, a residual error and a determination coefficient;
The step of generating the driving speed-acceleration amplification model according to the second historical data set, the overrun data distribution and the overrun data trend comprises taking the maximum speed, the acceleration and the speed corresponding to each line as variables and bringing the variables into the model parameters to obtain G Max ＝F(S Max ,G t ,S t ) Wherein G is Max For the maximum possible acceleration corresponding to the target line, S Max G is the maximum driving speed corresponding to the target line t For the corresponding acceleration of the target line, S t The corresponding speed of the target line.
By adopting the technical scheme, a more accurate driving speed-acceleration amplification model can be obtained, so that the collected driving speed and acceleration are converted during subsequent early warning;
the acceleration level and the acceleration overrun range can be set according to actual collection precision of collection equipment, for example, for a conventional vehicle shaking instrument, the acceleration level is divided into level I to level IV, for vertical acceleration, threshold values of level I to level IV are respectively 0.10m/s2, 0.15m/s2, 0.20m/s2 and 0.25m/s2, for horizontal acceleration, threshold values of level I to level IV are respectively 0.06m/s2, 0.10m/s2, 0.15m/s2 and 0.20m/s2, the acceleration overrun range can be 0.02m/s2, when an overrun range is defined, threshold values corresponding to level III can be used as overrun ranges, after the overrun range is defined, the overrun range can be screened in first history overrun data through the overrun range, and a second history data set can be obtained;
And fitting according to the second historical data set to obtain parameters forming the driving speed-acceleration amplification model, and then obtaining the corresponding amplification model by taking the maximum speed, the acceleration and the speed corresponding to each line as variables.
Optionally, performing density cluster analysis on the overrun result data set to obtain a first overrun data point cluster includes:
calculating an overrun point cluster according to a preset density threshold value and the disease radius;
inquiring all points in the overrun result data set according to the overrun point mileage range in the overrun point cluster, and calculating the proportion of the number of overrun points in the overrun point cluster to all overrun points in the section of the line;
comparing the ratio with a preset ratio threshold;
if the ratio is greater than or equal to the ratio threshold, taking the disease radius as a demarcation radius;
if the ratio is smaller than the ratio threshold, the disease radius is decreased according to a preset attenuation radius until the ratio is larger than or equal to the ratio threshold, and the decreased disease radius is used as a demarcation radius;
and acquiring a minimum milestone and a maximum milestone of the overrun points in the overrun point cluster according to the demarcation radius, and determining the milestone range of the disease point.
By adopting the technical scheme, the method not only can obtain the primary disease point range, but also can denoise the point clusters, and the density cluster analysis method can be a DBSCAN algorithm or an OPTICS algorithm;
the density threshold and the disease radius can be selected according to actual conditions, for example, when a DBSCAN algorithm is adopted, the disease radius is used as a neighborhood radius, for example, 200 meters, and the density threshold is set to be 5;
after the overrun point cluster is obtained, the proportion of the number of points in the overrun point cluster to the total number of points in the area is calculated, for example, the proportion can be 90%, namely, the number of points in the overrun point cluster is 90% of the total number of points in the area, the overrun point cluster is an available overrun point cluster, if the proportion is insufficient, the neighborhood radius is reduced according to the preset attenuation radius until the requirement is met, denoising is completed, in the application, each overrun point is positioned according to the mileage of the overrun point on the track, and therefore, after the available overrun point cluster is finally obtained, the starting point and the finishing point of the target road section can be obtained according to the maximum and minimum mileage marks.
Optionally, the step of performing development trend analysis on the first overrun data point cluster to obtain a second overrun data point cluster and disease levels corresponding to the second overrun data point cluster includes:
Judging that G is the same as G in the first overrun data point cluster Max Trend of time-series change of (3);
if the trend is to increase and then decrease, the diseases of the line are treated, the G Max The corresponding point cluster is a non-effective point cluster; if the trend remains to grow, the line is subject to disease, the G Max The corresponding point cluster is an effective point cluster, according to the G Max And a difference value of a preset overrun amplitude, dividing the G Max Grade of the corresponding disease.
By adopting the technical scheme, the overrun data point clusters are further screened to obtain a second overrun data point cluster and corresponding disease levels which need to be maintained in time, and the overrun threshold and overrun range before can be used as dividing basis when determining the disease levels, for example, the overrun threshold is 0.02m/s2 higher than the III level threshold and is low in risk, the overrun data point cluster is 0.02m/s2-0.04m/s2 higher than the III level threshold and is high in risk.
Optionally, acquiring second overrun data, and updating parameters of a density cluster analysis method and the driving speed-acceleration amplification model according to the second overrun data.
By adopting the technical scheme, after the operation and maintenance personnel conduct investigation and maintenance according to the early warning, actual investigation data are obtained, the actual investigation data are compared with the original data, second overrun data can be obtained, and the original model and method are optimized.
In a second aspect, the present application provides a high-speed railway line disease early warning device, which adopts the following technical scheme:
high-speed railway line disease early warning device includes:
the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring overrun data of a target train in a target time period, wherein the overrun data comprises acceleration information of the target train, and a work area, a line, a running class, a detection time, a mileage and a running speed corresponding to the acceleration information; the conversion module is used for inputting the overrun data into a pre-established running speed-acceleration amplification model to obtain an overrun result data set, wherein the overrun result data set is configured as a set of maximum acceleration reached by a target train when each line runs at the maximum running speed;
the first analysis module is used for performing density cluster analysis on the overrun result data set to obtain a first overrun data point cluster;
the second analysis module is used for carrying out development trend analysis on the first overrun data point cluster to obtain a second overrun data point cluster and disease levels corresponding to the second overrun data point cluster;
and the early warning module is used for carrying out early warning according to the second overrun data point cluster and the disease grade corresponding to the second overrun data point cluster.
In a third aspect, the present application provides a computer device, which adopts the following technical scheme:
a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of the first aspect when executing the program.
In a fourth aspect, the present application provides a computer readable storage medium, which adopts the following technical scheme:
a computer readable storage medium storing a computer program capable of being loaded by a processor and executing any one of the methods of the first aspect.
In summary, the present application includes at least one of the following beneficial technical effects:
the line disease is found more timely, and the data such as acceleration information and the like are analyzed by acquiring overrun data of a target train in a target time period to replace manual inspection depending on skylight time, so that a disease road section and a corresponding disease grade are automatically obtained;
the line disease position is positioned more accurately, and the range in which the disease possibly exists is reduced twice through density cluster analysis and development trend analysis, so that the operation and maintenance personnel can conveniently check and maintain;
the manual inspection cost is reduced, false inspection and missing inspection phenomena are reduced, and operation and maintenance personnel only need to go to the corresponding road section for inspection and maintenance according to the early-warning disease position and the disease grade.
Drawings
Fig. 1 is a schematic flow chart of a first embodiment of a method for warning of high-speed railway line diseases in the present application.
Fig. 2 is a second flow chart of a first embodiment of a method for warning of high-speed railway line diseases in the present application.
Fig. 3 is a third flow chart of a first embodiment of a method for warning of high-speed railway line diseases in the present application.
Fig. 4 is a schematic flow chart of a second embodiment of a method for warning of high-speed railway line diseases in the present application.
Fig. 5 is a block diagram of a high-speed railway line disease warning device in the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to fig. 1 to 5 and the embodiments. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The embodiment of the application discloses a first embodiment of a high-speed railway line disease early warning method.
Referring to fig. 1, the high-speed railway line disease early warning method includes:
s1, acquiring overrun data of a target train in a target time period, wherein the overrun data comprises acceleration information of the target train, and a work area, a line, a running category, detection time, mileage and running speed corresponding to the acceleration information;
According to the method, the track local irregularity analysis can be carried out by the acceleration information, wherein the acceleration information is the vibration acceleration of the target train and comprises horizontal vibration acceleration and vertical vibration acceleration, factors affecting the horizontal vibration acceleration of the target train comprise curves, continuous small directions of turnout areas, hard bending of steel rails, joint support mouths, poor track gauges, poor track gauge change rates, alternately uneven abrasion of straight sections of the steel rails and other diseases, factors affecting the vertical vibration acceleration of the target train comprise diseases such as poor track geometry, poor joint comprehensive state, serious or uneven track bed elasticity and the like, the horizontal vibration acceleration and the vertical vibration acceleration are generally divided into four grades according to numerical values, early warning is needed when approaching three grades, three grades and four grades, and the obtained acceleration information can be subjected to analysis in subsequent steps to early warning on the track disease conditions;
the specific position of the target train at a certain moment, namely the position of a certain section of track is analyzed in a local irregularity manner, wherein the work area can comprise a responsible unit and a unit number corresponding to the section of track, the line can comprise a line name and a line number corresponding to the section of track, the line respectively represents the running direction of the train on the section of track and comprises an ascending direction and a descending direction, the mileage represents the position of a mileage pile corresponding to the section of track, the running speed represents the running speed of the train on the section of track when data acquisition is performed, and the position of the target train and the disease degree of the section of track when the data acquisition is performed can be obtained more accurately by combining acceleration information, so that the development trend of the disease degree of the section of track is obtained; the target time period can be selected according to actual conditions, and can be a certain time period in daytime, the target train can be a plurality of trains running through a line to be early-warned, overrun data can be obtained through a vehicle-mounted vehicle shaking instrument on the target train, for example, through obtaining overrun data of 9:00 to 21:00, a relevant early-warning report is generated after analysis, the position of a possible disease point of a target road section is given, an operation and maintenance person is guided to carry out manual inspection in a window time more efficiently and purposefully, inspection efficiency is improved, and meanwhile, the data manually collected by the operation and maintenance person during inspection and the actual inspection result can be further optimized to an early-warning model.
S2, inputting overrun data into a pre-established running speed-acceleration amplification model to obtain an overrun result data set, wherein the overrun result data set is configured as a set of maximum accelerations reached by a target train when each line runs at the maximum running speed;
when the data is collected, the speed of the target train is not fixed, and may be lower in running speed, and the corresponding vibration acceleration is lower in running speed, so that it is difficult to reflect the defect condition of the track through the vibration acceleration, therefore, the collected running speed and vibration acceleration need to be converted to the maximum running speed and the corresponding maximum vibration acceleration through a running speed-acceleration amplification model, meanwhile, the data is more convenient to analyze after conversion, and of course, when the acceleration conversion is performed, the adopted data is not limited to the data in the target time period in the S1, for example, when the early warning is performed, the target time period is 9:00 to 21:00, but when the maximum running speed corresponding to the line is calculated, the overrun data in a longer time period can be obtained, such as the previous 7 days, the previous 5 days, and the like, so as to obtain the more accurate maximum running speed, and then the conversion obtains the more accurate maximum acceleration;
Referring to fig. 2, as a preferred implementation of the present embodiment, the step of creating the driving speed-acceleration amplification model includes: s201, acquiring first historical overrun data;
s202, calculating a data base line according to first historical overrun data to obtain a disease rule;
s203, preprocessing the first historical overrun data to obtain a second historical data set;
s204, obtaining overrun data distribution and overrun data trend according to the disease rules and the second historical data set;
s205, generating a driving speed-acceleration amplification model according to the overrun data distribution and the overrun data trend;
by adopting the implementation mode, an accurate running speed-acceleration amplification model can be established according to the historical overrun data, so that the acquired running speed and vibration acceleration are converted during early warning, and the maximum vibration acceleration corresponding to the running of the target train at the maximum running speed is obtained;
the first historical overrun data can be selected according to actual conditions, such as historical overrun data of the last year or two years, a data base line can be calculated by manually analyzing historical vehicle shaking instrument data and operation maintenance data distribution conditions, trends and rules and combining the actual conditions of rail diseases, and finally overrun data with a certain proportion in a certain specific radius with the actual disease point as a circle center is obtained and corresponds to one disease point, namely a disease rule;
The preprocessing is mainly to acquire data of vibration acceleration in a certain section in the first historical overrun data, such as data of vibration acceleration in the vicinity of three stages, and under the condition that the vibration acceleration is low when the vehicle runs at the highest speed, the road section can be judged to have low disease degree, the processing is not needed temporarily, the higher vibration acceleration usually corresponds to the serious disease degree and the probability of the serious degree to continue to develop is also high, so that after the preprocessing, a model can be constructed by more proper data; still further, the step of calculating a data baseline based on the first historical overrun data to obtain a disease rule includes:
obtaining a disease radius corresponding to a disease point according to the distribution condition, the trend and the rule of the historical vehicle shaking instrument data, the distribution condition, the trend and the rule of the operation maintenance data and the disease point data in the first historical overrun data;
the correlation between the running speed and the acceleration in the first historical overrun data is analyzed to obtain a running speed difference section, and the running speed difference section is configured to be the same as the acceleration corresponding to a plurality of running speeds with running speed differences in the running speed difference section;
Preprocessing the first historical overrun data to obtain a second historical data set, wherein the step of preprocessing the first historical overrun data comprises the steps of calculating an overrun interval according to a preset acceleration grade and an acceleration overrun range, and selecting data of overrun acceleration of each line in the overrun interval according to the overrun interval and the first historical overrun data to obtain the second historical data set;
the step of obtaining the overrun data distribution and overrun data trend according to the disease rule and the second historical data set comprises the steps of calculating model parameters corresponding to each line according to the disease rule and the second historical data set, and taking the model parameters as the overrun data distribution and overrun data trend, wherein the model parameters comprise slopes, intercepts, standard deviations of prediction errors, t statistics, residual errors and determination coefficients;
the step of generating a driving speed-acceleration amplification model according to the second historical data set, the overrun data distribution and the overrun data trend comprises the steps of taking the maximum speed, the acceleration and the speed corresponding to each line as variables and bringing the variables into model parameters to obtain G Max ＝F(S Max ,G t ,S t ) Wherein G is Max For the maximum possible acceleration corresponding to the target line, S Max G is the maximum driving speed corresponding to the target line t For the corresponding acceleration of the target line, S t The speed corresponding to the target line is set;
by adopting the embodiment, not only the disease radius corresponding to the more accurate disease point can be obtained according to the historical data and the density cluster analysis can be carried out in the subsequent step, but also the calculation amount in the process of establishing a model and subsequent early warning can be reduced through the driving speed difference section, and the early warning speed can be improved
The accurate driving speed-acceleration amplification model can be obtained, so that the collected driving speed and acceleration are converted in the follow-up early warning process;
the acceleration level and the acceleration overrun range can be set according to actual collection precision of collection equipment, for example, for a conventional vehicle shaking instrument, the acceleration level is divided into level I to level IV, for vertical acceleration, threshold values of level I to level IV are respectively 0.10m/s2, 0.15m/s2, 0.20m/s2 and 0.25m/s2, for horizontal acceleration, threshold values of level I to level IV are respectively 0.06m/s2, 0.10m/s2, 0.15m/s2 and 0.20m/s2, the acceleration overrun range can be 0.02m/s2, when an overrun range is defined, threshold values corresponding to level III can be used as overrun ranges, after the overrun range is defined, the overrun range can be screened in first history overrun data through the overrun range, and a second history data set can be obtained;
And fitting according to the second historical data set to obtain parameters forming the driving speed-acceleration amplification model, and then obtaining the corresponding amplification model by taking the maximum speed, the acceleration and the speed corresponding to each line as variables.
S3, performing density cluster analysis on the overrun result data set to obtain a first overrun data point cluster;
by adopting the embodiment, the approximate range of the disease points can be obtained, in general, one disease point can affect one section of track around the disease point, namely a plurality of overrun data points can be all caused by the same disease point, the approximate range of the disease points can be obtained by dividing the overrun data points in the preset neighborhood range into one point cluster, and operation and maintenance personnel can conduct investigation in the corresponding range; referring to fig. 3, as a preferred embodiment of the present application, the step of performing density cluster analysis on the overrun result dataset to obtain a first overrun data point cluster includes:
s301, calculating an overrun point cluster according to a preset density threshold value and a preset disease radius;
s302, calculating the proportion of overrun points, namely inquiring all points in overrun result data sets according to the mileage range of overrun points in overrun point clusters, and calculating the proportion of the number of overrun points in the overrun point clusters to all overrun points in the section of the line;
S303, comparing the proportion with a preset proportion threshold value, and judging whether the proportion is smaller than the proportion threshold value or not;
s304, taking the current disease radius as a demarcation radius, specifically taking the disease radius as the demarcation radius if the ratio is greater than or equal to a ratio threshold value;
s305, reducing the disease radius to obtain a demarcation radius, specifically, if the ratio is smaller than a ratio threshold value, reducing the disease radius according to a preset attenuation radius until the ratio is greater than or equal to the ratio threshold value, and taking the reduced disease radius as the demarcation radius; s306, acquiring a minimum milestone and a maximum milestone of an overrun point in the overrun point cluster according to the demarcation radius, and determining the milestone range of the disease point.
By adopting the embodiment, the primary disease point range can be obtained, the point clusters can be denoised, and the density cluster analysis method can be a DBSCAN algorithm or an OPTICS algorithm;
the density threshold and the disease radius can be selected according to actual conditions, in the embodiment, a DBSCAN algorithm is adopted, the disease radius is used as a neighborhood radius, the neighborhood radius is 200 meters, and the density threshold is set to be 5;
the DBSCAN algorithm is fully called Density-Based SpatialClustering of Applications with Noise, namely: a spatial clustering algorithm robust to noise based on density, a DBSCAN algorithm has the following advantages:
The DBSCAN can find cluster clusters with random shapes, unlike the traditional distance-based clustering method (such as K-means), the DBSCAN can effectively find cluster clusters with irregular shapes because the cluster clusters are clustered based on density rather than distance;
the sensitivity to parameters is low, compared with some methods (such as K mean value) requiring the pre-specified clustering number, the DBSCAN algorithm is relatively insensitive to the selection of the density threshold epsilon and the data point number MinPts in the minimum neighborhood, so that the DBSCAN is easier to tune in practical application, and is more suitable for the situations of uneven data distribution and large density change;
abnormal points can be identified, and the DBSCAN can effectively mark the isolated points or the abnormal points which do not belong to any cluster as noise points, so that abnormal data can be identified and filtered, and the clustering result is purer;
the DBSCAN does not need to specify the number of clusters in the data set in advance, so that the DBSCAN has more flexibility in practical application, especially when the feature knowledge of the data set is limited or the property of the data set is uncertain;
for a high-speed railway, the reasons for causing diseases are various, disease points are distributed along the line direction, the generated cluster, namely the shape of the overrun cluster, is quite irregular, meanwhile, the line distance is long, the number of clusters is difficult to be appointed in advance, more anomalies and noise are often generated due to the complex environment where the track is located, and therefore, the DBSCAN algorithm is very suitable for disease early warning of the high-speed railway;
After the overrun point cluster is obtained, the proportion of the number of points in the overrun point cluster to the total number of points in the area is calculated, for example, the proportion can be 90%, namely, the number of points in the overrun point cluster is 90% of the total number of points in the area, the overrun point cluster is an available overrun point cluster, if the proportion is insufficient, the neighborhood radius is reduced according to the preset attenuation radius until the requirement is met, denoising is completed, in the application, each overrun point is positioned according to the mileage of the overrun point on the track, and therefore, after the available overrun point cluster is finally obtained, the starting point and the finishing point of the target road section can be obtained according to the maximum and minimum mileage marks.
S4, carrying out development trend analysis on the first overrun data point cluster to obtain a second overrun data point cluster and disease grades corresponding to the second overrun data point cluster;
the development trend is mainly the development trend of vibration acceleration, the trend is different, may correspond to different diseases, diseases have been processed, disease serious conditions are different, etc., can also obtain the correspondent disease grade according to the concrete numerical value change condition of vibration acceleration, after excluding the conditions that the diseases have been processed, disease severity is lower, etc., the data point cluster of overrun obtained is the second overrun data point cluster;
As a preferred embodiment, the step of performing a trend analysis on the first overrun data point cluster to obtain the second overrun data point cluster and the disease level corresponding to the second overrun data point cluster includes:
judging G in the first overrun data point cluster Max Trend of time-series change of (3);
if the trend is to increase and decrease, the diseases of the line are treated, G Max The corresponding point cluster is a non-effective point cluster;
if the trend remains to grow, the diseases of the line are to be treated, G Max The corresponding point cluster is an effective point cluster, according to G Max And a difference value of a preset overrun amplitude, dividing G Max The grade of the corresponding disease;
by adopting the embodiment, the overrun data point cluster is further screened to obtain a second overrun data point cluster and a corresponding disease grade which need to be maintained in time, and the overrun threshold and the overrun range before the overrun data point cluster and the corresponding disease grade are used as dividing basis when the disease grade is determined, for example, the overrun threshold is 0.02m/s2 higher than the III level threshold and is a low risk, and the overrun data point cluster is 0.02m/s2-0.04m/s2 higher than the III level threshold and is a high risk.
S5, early warning is carried out according to the second overrun data point cluster and the disease grade corresponding to the second overrun data point cluster;
The second overrun data point cluster can obtain a more accurate disease point range and severity, and can guide operation and maintenance personnel to go to the investigation treatment in time.
To sum up, the beneficial effects of the embodiment are as follows:
the line disease is found more timely, and the data such as acceleration information and the like are analyzed by acquiring overrun data of a target train in a target time period to replace manual inspection depending on skylight time, so that a disease road section and a corresponding disease grade are automatically obtained;
the line disease position is positioned more accurately, and the range in which the disease possibly exists is reduced twice through density cluster analysis and development trend analysis, so that the operation and maintenance personnel can conveniently check and maintain;
the manual inspection cost is reduced, false inspection and missing inspection phenomena are reduced, and operation and maintenance personnel only need to go to the corresponding road section for inspection and maintenance according to the early-warning disease position and disease grade;
further comprises:
and constructing a reasonable driving speed-acceleration amplification model and adopting a DBSCAN algorithm which is more suitable for a high-speed railway, so that the positioning accuracy of the disease position is further improved.
The application also discloses a second embodiment of the high-speed railway line disease early warning method.
Referring to fig. 4, compared to the first embodiment, in addition to step S1 to step S5, the present embodiment further includes:
S6, acquiring second overrun data, and updating parameters of a density cluster analysis method and a driving speed-acceleration amplification model according to the second overrun data;
by adopting the implementation mode, after the operation and maintenance personnel conduct investigation and maintenance according to the early warning, actual investigation data are obtained, and after the actual investigation data are compared with the original data, second overrun data can be obtained, and the original model and method are optimized.
The beneficial effects of this embodiment also include: and updating parameters of the model and the algorithm in time, and further improving the positioning accuracy of the disease position.
The embodiment of the application also discloses a high-speed railway line disease early warning device.
Referring to fig. 5, the high-speed railway line defect warning apparatus includes:
the acquisition module is used for acquiring overrun data of the target train in the target time period, wherein the overrun data comprises acceleration information of the target train, and work areas, lines, running categories, detection time, mileage and running speed corresponding to the acceleration information;
the conversion module is used for inputting the overrun data into a pre-established running speed-acceleration amplification model to obtain an overrun result data set, wherein the overrun result data set is configured as a set of maximum acceleration reached by a target train when each line runs at the maximum running speed;
The first analysis module is used for performing density cluster analysis on the overrun result data set to obtain a first overrun data point cluster;
the second analysis module is used for carrying out development trend analysis on the first overrun data point cluster to obtain a second overrun data point cluster and disease levels corresponding to the second overrun data point cluster;
and the early warning module is used for carrying out early warning according to the second overrun data point cluster and the disease grade corresponding to the second overrun data point cluster.
The high-speed railway line disease early warning device can realize any one of the high-speed railway line disease early warning methods, and the specific working process of each module in the high-speed railway line disease early warning device can refer to the corresponding process in the method embodiment.
In several embodiments provided herein, it should be understood that the provided methods and apparatus may be implemented in other ways. For example, the device embodiments described above are merely illustrative; for example, a division of a module is merely a logical function division, and there may be another division manner in actual implementation, for example, multiple modules may be combined or may be integrated into another device, or some features may be omitted or not performed.
The embodiment of the application also discloses a computer device.
The computer equipment comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, and the processor realizes the high-speed railway line disease early warning method when executing the computer program.
The embodiment of the application also discloses a computer readable storage medium.
A computer-readable storage medium storing a computer program that can be loaded by a processor and that performs any one of the above-described high-speed railway line disease warning methods.
Wherein the computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution apparatus, device, or apparatus; program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
In the foregoing embodiments, the descriptions of the embodiments are focused on, and for those portions of one embodiment that are not described in detail, reference may be made to the related descriptions of other embodiments.
The foregoing description of the preferred embodiments of the present application is not intended to limit the scope of the application, in which any feature disclosed in this specification (including abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. That is, each feature is one example only of a generic series of equivalent or similar features, unless expressly stated otherwise.
Claims (10)
1. The high-speed railway line disease early warning method is characterized by comprising the following steps of:
acquiring overrun data of a target train in a target time period, wherein the overrun data comprises acceleration information of the target train, and a work area, a line, a running category, detection time, mileage and running speed corresponding to the acceleration information;
inputting the overrun data into a pre-established driving speed-acceleration amplification model to obtain an overrun result data set, wherein the overrun result data set is configured as a set of maximum acceleration reached by a target train when each line runs at the maximum driving speed;
performing density cluster analysis on the overrun result data set to obtain a first overrun data point cluster;
performing development trend analysis on the first overrun data point cluster to obtain a second overrun data point cluster and disease grades corresponding to the second overrun data point cluster;
and carrying out early warning according to the second overrun data point cluster and the disease grade corresponding to the second overrun data point cluster.
2. The high-speed railway line disease warning method according to claim 1, wherein the step of establishing the running speed-acceleration amplification model includes:
Acquiring first historical overrun data;
calculating a data base line according to the first historical overrun data to obtain a disease rule;
preprocessing the first historical overrun data to obtain a second historical data set;
obtaining overrun data distribution and overrun data trend according to the disease rules and the second historical data set;
and generating the driving speed-acceleration amplification model according to the overrun data distribution and the overrun data trend.
3. The method for early warning of a high-speed railway line fault according to claim 2, wherein the step of calculating a data base line according to the first historical overrun data to obtain a fault rule comprises:
obtaining a disease radius corresponding to a disease point according to the distribution condition, the trend and the rule of the historical vehicle shaking instrument data, the distribution condition, the trend and the rule of the operation maintenance data and the disease point data in the first historical overrun data;
and analyzing the correlation between the running speed and the acceleration in the first historical overrun data to obtain a running speed difference section, wherein the running speed difference section is configured such that the acceleration corresponding to a plurality of running speeds with running speed differences within the running speed difference section is the same.
4. The high-speed railway line disease early warning method according to claim 3, characterized in that:
the step of preprocessing the first historical overrun data to obtain a second historical data set comprises the steps of calculating an overrun interval according to a preset acceleration grade and an acceleration overrun range, and selecting data of overrun acceleration of each line in the overrun interval according to the overrun interval and the first historical overrun data to obtain the second historical data set;
the step of obtaining the overrun data distribution and the overrun data trend according to the disease rule and the second historical data set comprises the steps of calculating model parameters corresponding to each line according to the disease rule and the second historical data set, and taking the model parameters as the overrun data distribution and the overrun data trend, wherein the model parameters comprise a slope, an intercept, a standard deviation of a prediction error, a t statistic, a residual error and a determination coefficient;
the step of generating the driving speed-acceleration amplification model according to the second historical data set, the overrun data distribution and the overrun data trend comprises taking the maximum speed, the acceleration and the speed corresponding to each line as variables and bringing the variables into the model parameters to obtain G Max ＝F(S Max ,G t ,S t ) Wherein G is Max For the maximum possible acceleration corresponding to the target line，S Max G is the maximum driving speed corresponding to the target line t For the corresponding acceleration of the target line, S t The corresponding speed of the target line.
5. The method for early warning of high-speed railway line diseases according to claim 4, wherein the step of performing density cluster analysis on the overrun result data set to obtain a first overrun data point cluster comprises:
calculating an overrun point cluster according to a preset density threshold value and the disease radius;
inquiring all points in the overrun result data set according to the overrun point mileage range in the overrun point cluster, and calculating the proportion of the number of overrun points in the overrun point cluster to all overrun points in the section of the line;
comparing the ratio with a preset ratio threshold;
if the ratio is greater than or equal to the ratio threshold, taking the disease radius as a demarcation radius;
if the ratio is smaller than the ratio threshold, the disease radius is decreased according to a preset attenuation radius until the ratio is larger than or equal to the ratio threshold, and the decreased disease radius is used as a demarcation radius;
and acquiring a minimum milestone and a maximum milestone of the overrun points in the overrun point cluster according to the demarcation radius, and determining the milestone range of the disease point.
6. The method for early warning of high-speed railway line diseases according to claim 5, wherein the step of performing development trend analysis on the first overrun data point cluster to obtain a second overrun data point cluster and disease levels corresponding to the second overrun data point cluster comprises:
judging that G is the same as G in the first overrun data point cluster Max Trend of time-series change of (3);
if the trend is to increase and then decrease, the diseases of the line are treated, the G Max The corresponding point cluster is a non-effective point cluster; if it isThe trend keeps growing, then the diseases of the line are to be treated, the G Max The corresponding point cluster is an effective point cluster, according to the G Max And a difference value of a preset overrun amplitude, dividing the G Max Grade of the corresponding disease.
7. The high-speed railway line disease warning method according to any one of claims 1 to 6, characterized by further comprising: and acquiring second overrun data, and updating parameters of a density cluster analysis method and the driving speed-acceleration amplification model according to the second overrun data.
8. The utility model provides a high-speed railway circuit disease early warning device which characterized in that includes:
the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring overrun data of a target train in a target time period, wherein the overrun data comprises acceleration information of the target train, and a work area, a line, a running class, a detection time, a mileage and a running speed corresponding to the acceleration information;
The conversion module is used for inputting the overrun data into a pre-established running speed-acceleration amplification model to obtain an overrun result data set, wherein the overrun result data set is configured as a set of maximum acceleration reached by a target train when each line runs at the maximum running speed;
the first analysis module is used for performing density cluster analysis on the overrun result data set to obtain a first overrun data point cluster;
the second analysis module is used for carrying out development trend analysis on the first overrun data point cluster to obtain a second overrun data point cluster and disease levels corresponding to the second overrun data point cluster;
and the early warning module is used for carrying out early warning according to the second overrun data point cluster and the disease grade corresponding to the second overrun data point cluster.
9. A computer device, characterized by: comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the high-speed railway line fault warning method according to any one of claims 1 to 7 when the program is executed.
10. A computer-readable storage medium, characterized by: a computer program capable of being loaded by a processor and executing the high-speed railway line disease warning method according to any one of claims 1 to 7 is stored.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311565787.1A CN117755354A (en) | 2023-11-22 | 2023-11-22 | High-speed railway line disease early warning method and device and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311565787.1A CN117755354A (en) | 2023-11-22 | 2023-11-22 | High-speed railway line disease early warning method and device and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117755354A true CN117755354A (en) | 2024-03-26 |
Family
ID=90320870
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311565787.1A Pending CN117755354A (en) | 2023-11-22 | 2023-11-22 | High-speed railway line disease early warning method and device and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117755354A (en) |
-
2023
- 2023-11-22 CN CN202311565787.1A patent/CN117755354A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020244288A1 (en) | Method and apparatus for evaluating truck driving behaviour based on gps trajectory data | |
| CN102887159B (en) | Method and device for railway vehicle tracking and planned reporting | |
| CN108595374B (en) | Method and device for identifying geometric tiny changes of high-speed railway track | |
| CN113420367B (en) | Subway steel rail wave grinding detection method and device based on vibration and noise response | |
| WO2021001219A1 (en) | Monitoring, predicting and maintaining the condition of railroad elements with digital twins | |
| CN110736999B (en) | Railway turnout detection method based on laser radar | |
| CN106218668B (en) | Wheel out of round degree detection method and device | |
| Xu et al. | Optimizing the alignment of inspection data from track geometry cars | |
| CN113947130B (en) | Training and using method and device for regression prediction AI model of polygonal wear waveform of wheel | |
| Braga et al. | Multivariate statistical aggregation and dimensionality reduction techniques to improve monitoring and maintenance in railways: The wheelset component | |
| Hoelzl et al. | On-board monitoring for smart assessment of railway infrastructure: A systematic review | |
| CN110503209B (en) | Steel rail analysis early warning model construction method and system based on big data | |
| CN110254478B (en) | Roadbed deformation disease identification method and device | |
| Men et al. | Health assessment of high-speed train wheels based on group-profile data | |
| CN104457644B (en) | The detection method and device of non-pulse abnormal data in track geometry detection data | |
| CN117755354A (en) | High-speed railway line disease early warning method and device and computer equipment | |
| CN110555188A (en) | analysis method for wheel parameter abrasion trend of motor train unit | |
| CN116681912A (en) | Rail gauge detection method and device for railway turnout | |
| CN113609565B (en) | Acceptance method and device for static long wave irregularity of railway bridge track | |
| Zhu et al. | Evaluating the wheelset health status of rail transit vehicles: Synthesis of wear mechanism and data-driven analysis | |
| CN115372932A (en) | System and method for evaluating driveability of monorail crane and predicting roadway risk | |
| CN113569491A (en) | Analysis and correction method and device for wheel set size detection data | |
| CN113032907A (en) | Method and system for correcting vehicle shaking disease data deviation based on waveform correlation | |
| CN111444658A (en) | Groove rail geometric parameter trend prediction method and system | |
| CN117725804B (en) | Rail geometrical parameter and vehicle dynamics fusion influence analysis method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |